Block a user
d05205d1eb
Add a hunter skill: a sweep and a diff review are different jobs with different output contracts
c801851c66
Correct the isLoopback javadoc: after #305 narrowing this range refuses a caller, it does not promote one
#317: refuse an unresolved caller instead of promoting it to primary
A failed peer-PID lookup promotes a worker to primary: the escalation PaneLocator's own javadoc names, through a trigger CB-161 did not close
A failed peer-PID lookup promotes a worker to primary: the escalation PaneLocator's own javadoc names, through a trigger CB-161 did not close
Merged as de70aa3. Pushed to main.
What I checked myself
Build after the merge, unpiped, on main with #315 already in it:
Tests run: 1330, Failures: 0, Errors: 0, Skipped: 0 —…
#315: FixedPlacementPolicy now honors the retry loop's unreachable set
The
fixed placement policy ignores the retry loop's unreachable set, so an unqualified spawn retries the same dead profile and never tries the healthy one
The
fixed placement policy ignores the retry loop's unreachable set, so an unqualified spawn retries the same dead profile and never tries the healthy one
Merged to main in 77ad886.
My alternative mechanism was wrong, and the worker was right to reject it
I offered two candidates and said the second was attractive because it "makes the loop…
A reply delivered while release() is running is stranded unacked forever: #298 closed "already held", not "arrives during release"
A failed peer-PID lookup promotes a worker to primary: the escalation PaneLocator's own javadoc names, through a trigger CB-161 did not close
The dirty-worktree check runs before the worker is stopped, so work written during teardown is deleted with no preserve and no snapshot
The
fixed placement policy ignores the retry loop's unreachable set, so an unqualified spawn retries the same dead profile and never tries the healthy one
#307: an ask() timeout no longer strands the worker's real reply
A worker's report is stranded when its fleet_ask timed out, and the lead is told the worker never replied
A worker's report is stranded when its fleet_ask timed out, and the lead is told the worker never replied
Merged to main in b9c2cf6. This is the one in the batch I judged most likely to have already cost us real work.
What I checked myself
The worker did not use the candidate mechanism…
b9c2cf69f4
Merge #307: a worker's real reply after an ask timeout completes its ticket instead of stranding
#309: clean partial worktrees after add failure
#308: refuse a spawn once the shutdown drain has started, and sweep stragglers